Research Challenges in PV reliability

RESEARCH CHALLENGES IN PV RELIABILITY
www.etip-pv.eu
List of Authors
Harry Wirth (Fh-ISE)
David Moser (Eurac Research)
Marko Topic (University of Ljubljana)
Mike van Iseghem (EDF)
Ulrike Jahn (TÜV),
Giovanna Adinolfi, Giorgio Graditi (ENEA)
Brian Azzopardi (MCAST)
Eszter Voroshazi (IMEC)
Wim Sinke (TNO)
Paula Mints (SPV Market Research)
Ralph Gottschalg (Fh-CSP)
Mauricio Richter (3E)
Guillermo Oviedo Hernandez, Paolo Chiantore (Baywa Italy)
Arnulf Jaeger-Waldau (JRC)
Veronica Bermudez (QEERI- Qatar Environment & Energy Research Institute)
Christoph Mayr (AIT)
Table of Contents
RESEARCH CHALLENGES IN PV RELIABILITY . . . . . . . . . . . . . . 5
1. WHY DO WE NEED ONGOING RESEARCH? . . . . . . . . . . . . . . 6
1.1 Huge PV deployment volumes and bankable investments. . . . . . . . 6
1.2 Ongoing technology innovation. . . . . . . . . . . . . . . . . . . . . . 7
2. WHAT TOPICS NEED TO BE COVERED? . . . . . . . . . . . . . . . . 8
2.1 Advancement of type testing and quality check procedures. . . . . . . 8
2.2 Holistic approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 PV plant design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4 O&M Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Layout and Printing
Secretariat of the
European Technology and Innovation Platform for Photovoltaics
Tel: +49-89-720 12 722
Fax: +49-89-720 12 791
[email protected]
Disclaimer
The opinions expressed in this document are the sole responsibility of the European Photovoltaic Technology
and Innovation Platform and do not necessarily represent the official position of the European Commission.
“The project has received funding from the European Union’s
Horizon 2020 research and innovation programme under
grant agreement No 825669”
2
3
RESEARCH CHALLENGES IN PV RELIABILITY
RESEARCH CHALLENGES IN PV RELIABILITY
Reliable solar PV power plants will deliver the expected performance throughout their entire service life.
They will experience very limited component degradation, failures or system down-time. Improving reliability
means derisking electricity production and return on investment of PV systems. Mitigation of risks supports
the bankability of PV systems and serves as an enabler for accelerated deployment of solar PV technologies.
Ongoing basic research remains necessary for the photovoltaic industry to continue advancing its knowledge
and improving lifetime system performance. This document addresses the foundations and reasons why
reliability research is crucial to the solar PV industry’s continued growth.
After a 6-year decline of the solar photovoltaic market in the European Union, solar deployment started
to grow again in 2018. This development is driven by the push towards a faster greenhouse gas reduction
needed to achieve the targets of the Paris Agreement. As a consequence, the European Union finally adopted
the recast of the Renewable Energy Directive in December 20181 and the new (2019 – 2024) Commission
in December 2019 launched an European Green Deal2 to achieve carbon neutrality by 2050 to follow and
implement the declaration of the European Parliament on a climate emergency on 28th November 2019.
By 2050, when the world must have achieved carbon neutrality, the European Technology and Innovation
Platform Photovoltaics (ETIP PV) forecasted in the ETIP PV vision for PV3 that solar PV will supply 70% of
the world’s electricity generation. We will use solar PV to move, heat, cool, drive eco-friendly industrial
processes and produce fuels. Solar PV will be deployed everywhere, in systems large and small, designed
for productivity and aesthetics as required.
To enable the needed rapid deployment of solar photovoltaics in Europe towards a TW level in the next
decade, basic research remains necessary so that the industry can continue advancing its knowledge and
improving system performance and lifetime. Basic research, however, is not sufficient to ensure progress
and assure quality. Reliability research (typically situated at higher TRL levels) is necessary to ensure that
PV power plants deliver the expected performance throughout their entire service life. Ongoing reliability
research will ensure that solar plants experience limited component degradation, failures, or system downtime.
Directive (EU) 2018/2001 of the European Parliament and of the Council of 11 December 2018 on the promotion of
the use of energy from renewable sources (recast), Official Journal of the European Union, L 328/82, 21.12.2018
2
European Commission Communication, The European Green Deal, COM (2019) 640final, 11 December 2019
3
https://etip-pv.eu/publications/etip-pv-publications/download/photovoltaic-solar-energy-big-and-beyond, Dec 2018
1
4
5
1. WHY DO WE NEED ONGOING RESEARCH?
1. WHY DO WE NEED ONGOING RESEARCH?
1. WHY DO WE NEED ONGOING RESEARCH?
1.1
Huge PV deployment volumes and bankable investments
Climate is being destabilized by greenhouse gas
emissions, bringing with it more extreme weather
events4. The unexpected extreme weather of today
will be commonplace tomorrow, challenging the
reliability of PV components. Europe is undergoing
an energy transition, switching from fossil fuels and
nuclear power to wind and solar PV as the primary
source for electricity generation. In the coming
decades, installed PV capacity is expected to rise
from 130 GW to the scale of 1 TW. Deployment of
PV assets on a terawatt scale requires a significant
financial commitment in the range of hundreds of
billions euros from utilities, electricity prosumers,
energy communities and governments. Given
the significant levels of deployment required,
and the speed at which the transition is taking
place, reliability research must ramp up alongside
deployment to ensure stable energy supply at low
cost, justifying and assuring bankability of deployed
PV systems and PV technology as a whole.
Levelized cost of electricity (LCOE) remains an
essential metric in the solar PV industry (as in all
other energy sector industries), often used to assess
progress in cost, and useful in setting expectations.
LCOE not only depends on investments (CAPEX), it
also includes operation and maintenance (O&M)
costs, and needs the utilization rate (production
in kWh/year) as input, which in turn is impacted
by service life, PV plant availability and annual
degradation rates of the installed PV equipment.
Globally, seventy-nine percent of solar modules
are manufactured in China and the countries of
Southeast Asia. A significant reliance on imports
requires good procedures for quality assurance of
components, appropriate to the environment in
which they will be installed.
1.2
Ongoing technology innovation
PV technology has been undergoing highly dynamic
innovation processes for LCOE reduction, which are
expected to continue over the next decades. The PV
sector is continuously struggling to improve device
efficiency and to reduce material and manufacturing
cost. Special attention is paid to reducing material
consumption (e.g. silicon, silver) and should be
paid to replace hazardous materials (e.g. lead) by
safe alternatives. Bifacial module technology is
quickly entering the market, requiring transparent
rear covers and a close look at rear side irradiance.
Highly efficient, yet temperature sensitive solar
cells call for new, low temperature interconnection
technologies. Innovation is also ongoing in the field
of PV mounting systems and plant design, e.g. to
accomodate bifacial modules or to design agrivoltaic
or floating systems. While many innovations follow
an incremental approach, disruptive technologies
such as tandem devices are being investigated
under the assumption that they will substantially
exceed single-junction Si efficiencies.
Although historic PV module designs have
demonstrated service life potentials beyond 35
years, the sector has learned that every innovation
step generates unforeseen reliability risks, not
detected by state-of-the-art testing and certification
procedures. In the context of innovative materials, it
is essential to avoid only screening for known failure
modes, that is, those derived from previously used
materials, and which do not necessarily cover the
failures that occur in new materials. Examples of
novel degradation mechanisms originating from
new technology or materials include PID, LID, LeTID,
back-sheet chalking, and thin-film metastabilities.
4
6
European PV power plant operators remain uncertain
about the materials’ quality and the properties
needed to ensure it. Though Bill-Of-Materials (BOMs)
are becoming more available from manufacturers,
these instruments do not provide information about
the quality of the components and their link with
the total reliability of the device. A small fraction
of the solar modules received by the European
operators from their Asian suppliers include innate
defects due to the choice of materials (polymers in
particular), which reduce module performance and
safety after the first 5-10 years of plant operation.
Many European utilities have recorded these defects.
These defects impact the performance and service
life of the modules and diminish solar electricity
production.
Beyond traditional ground-mounted or roof-mounted
PV power plants, PV integration into buildings and
vehicles is becoming more common. Integrated PV
applications include building facades, noise barriers,
road surfaces, and floating PV. Integrated applications
require unique module designs or materials, and
certification and test methods adapted to the new
functionalities and environmental challenges. Safety
and reliability are crucial for PV modules used in
integrated applications.
Novel tools and characterization techniques
allow more comprehensive approaches to quality assurance in operation and maintenance of PV
power plants. The deployment of cost-effective IoT
sensors enable module or string level monitoring.
Unmanned aerial vehicles combined with multispectral analysis provide complete views of large
power plants with high spacial resolution.
IPCC, 2018: Summary for Policymakers. In: Global Warming of 1.5°C. An IPCC Special Report on the impacts of
global warming of 1.5°C above pre-industrial levels and related global greenhouse gas emission pathways, in
the context of strength-ening the global response to the threat of climate change, sustainable development, and
efforts to eradicate poverty [Masson-Delmotte, V., P. Zhai, H.-O. Pörtner, D. Roberts, J. Skea, P.R. Shukla, A. Pirani,
W. Moufouma-Okia, C. Péan, R. Pidcock, S. Connors, J.B.R. Matthews, Y. Chen, X. Zhou, M.I. Gomis, E. Lonnoy, T.
Maycock, M. Tignor, and T. Waterfield (eds.)].
7
2. WHAT TOPICS NEED TO BE COVERED?
2. WHAT TOPICS NEED TO BE COVERED?
2.1
Advancement of type testing and quality check procedures
Type testing standards like IEC 61215 and IEC
61730 are based on experience from former or
incumbent technologies. Reliability Research is
required for their adaptation to novel and future
technologies, including heterojunction, passivated
contacts or dual junction cell devices, as well as
multiwire and shingle interconnection technologies.
Reliability Research helps to identify related novel
degradation mechanisms, to adapt test schemes
correspondingly, and to improve product design.
Increasing the knowledge of the properties of new
materials and designs with their link to module
quality is an important research item and will
support overall reliability.
Module type qualification mostly relies on separate
tests for different stresses which may not reveal all
relevant failure modes. This deficiency is especially
critical in case of highly innovative materials and
designs. Testing real-size modules in combined stress
chambers for novel failure modes is a significant
improvement of current practices. Reliability research
has to investigate the acceleration potential of
combined stress testing and to finally translate the
relevant stresses into more simple and efficient
procedures. Combined stress protocols require
collaborative research and knowledge sharing with
other industries.
Highly accelerated stress testing (HAST) procedures
are required to support not only technology
development, where it reduces cost and time-tomarket, but also for quick quality checks by product
off-takers. Reliability Research helps to understand
the options and limits of test acceleration.
For the global photovoltaic industry to address the
harsher environments of the future, it needs to
understand which system components, installation
techniques, and O&M practices are the most
appropriate. Climatic loads associated with
increasingly extreme weather conditions, especially
8
wind speeds and hail, need to be investigated and
translated to more demanding module type testing
procedures.
Although laboratory record cells efficiencies have
been continuously increasing in the last decades
together with commercial modules efficiencies, field
performance has not always been warranted and
performance data has rarely been shared with the
scientific community. Cost pressure might lead to
the development of tests in isolation of use cases
which might result in meaningless test sequences
for irrelevant failure mechanisms. Accelerated
degradation tests on component level need to
walk hand-in-hand with performance assessment
under relevant climatic conditions, and system level
testing including increasingly harsh environments.
Soiling, salinity, humidity, high temperatures and
high UV exposure may affect different technologies
in different ways and accelerate module degradation.
Field reliability testing is especially crucial for new
concepts and devices where we have little or no
experience in yearly performance ratio and extended
reliability. Incomplete design verification increases
risks of PV plant underperformance significantly.
Field reliability research based on high-quality
monitoring data is leading to a better understanding
of the degradation rates and failure modes leading
to the development of preventive maintenance
protocols and lower O&M costs, better future control
of costs, and thus more accurate LCOE prediction.
The reliability of power electronic devices like DC/
AC, DC/DC converters and power optimizers as
part of PV power plants is crucial. Conceptually,
failure rates can be estimated by field data, test
data, stress-strength analysis and/or combination
of the three methods. For an item operating for
a long time, the failure data can be collected and
categorized following potential failure mechanisms,
and hence, a complete reliability model for a
specific operating condition can be provided. This
approach requires long term operation of an item
2. WHAT TOPICS NEED TO BE COVERED?
and accurate categorizing the failure cause and
mechanisms under given operating conditions. [1]
Also, the IEC 61709 published in 2017 [2] provides a
guideline to use the failure rate data for predicting
the reliability of components considering the
temperature cycling in failure rate prediction
throughout an annual mission profile. [1] proposes
a general reliability prediction approach for power
converters according to the failure causes on the
most failure prone components.
Reliability models can be used for decision making
in reliable design, maintenance scheduling, systemlevel planning, as well as comparing different
alternatives such as topologies, control system,
modulation scheme, fault tolerant approaches and
so on. Finally, different operating and environmental
2.2
conditions have an impact on failure rates as well
as it causes constant failures or wear-out failures.
Novel applications in the context of PV integration
come along with specific stress effects. As an
example, vehicle, noise barrier and road integration
environment subject modules to severe levels of
dynamic mechanical stress, while floating PV exposes
modules and BOS components to high levels of
humidity. For building integration on the other hand,
module service life needs to safely achieve levels
typical for building materials. Reliability Research
needs to investigate the associated material stress
and to design and validate adequate accelerated
test procedures, based on significant degradation
indicators.
Holistic approaches
Today reliability research is often focused on isolated components and often limited to type approval testing.
This is somewhat different to other industries where reliability research is ongoing through the lifetime of
a product (Figure 1).
Figure 1: General Industry Approach to Reliability
[1]
Reliability Modeling of Power Electronic Converters: A General Approach, 2019,
https://ieeexplore.ieee.org/document/8769685
[2]
“IEC 61709 (2017): Electric Components - Reliability - Reference Conditions for Failure Rates and
Stress Models for Conversion”, 2017.
9
2. WHAT TOPICS NEED TO BE COVERED?
Each of the stages is associated with some research challenges. However, a reliable system can only be
addressed if a comprehensive quality process is followed. This is clear when considering a proven design (e.g.
according to IEC61215/61730) which is not manufactured appropriately or mishandled during installation –
neither will result in a reliable product. A fair share of current quality assurance is based on opinion, while
the industry requires consistent, science-based processes. It needs to address not only the PV module, but
also the other balance of system (BOS) components that affect PV plant performance and safety.
2.3
PV plant design
In a PV project the cost of correcting a failure
increases exponentially by each step along the value
chain from project development to plant operation.
Failure prevention instead of correction should
be considered as a first mitigation option with an
effective risk management strategy during system
design and planning. These mitigation options
will have costs which need to be counterbalanced
by a decrease in operational expenses. Typically,
during the design of a PV project, a component
qualification process is put in place. This contains
compatibility check, risk analysis, supplier audit,
and lesson learned.
Mitigation measures must be identified along the PV
value chain and assigned to various technical risks.
Typical mitigation measures during the design phase
are linked to component selection, O&M friendly
design, monitoring system design, accessibility of
the site, choice of inverter layout and quality check
of the input data for yield assessment.
The main research challenges associated to PV plant
design are related to (i) the need to create digital
twins of PV plants which can be then transferred to
the O&M phase, (ii) the introduction of disruptive
engineering concepts based on LCOE optimized
design, (iii) the reduction of uncertainties related
to yield assessments especially in novel emerging
PV module technologies.
10
2.4
O&M Support
As the European PV fleet gets older (10+ years), O&M
contractors are becoming the main risk managers
of PV plants aging and underperformance issues.
Field reliability assessment in the operation phase
of PV plants for 20-25 years after commissioning
has been mainly driven in the past by empirical
approaches and field experience. Furthermore,
laboratory tests that support field inspections come
at very high costs. These costs increase short-term
OPEX substantially and are usually prohibitive for
O&M contractors in a very competitive market.
In general terms, Reliability Research should be
aimed at minimizing or optimizing field operations
for lowering OPEX while improving remote failure
detection and troubleshooting for increasing
energy yield. The digitalization of the entire O&M
ecosystem should be at the center of the research
challenges, where system simulation (digital twin
concept), advanced diagnostic tools and intraday
yield forecasting would bridge the gap between the
ever-increasing penetration of renewable resources
and the grid. Grid integration approaches, such as
smart grids, storage, and electric mobility, together
with a fully digitalized O&M sector, would pave the
path for the energy transition.
Artificial Intelligence (AI) techniques, such as
Machine Learning, have started to be applied to the
huge amount of monitoring data already available
(now considered big data), opening new ways for
forecasting approaches, supporting preventive
2. WHAT TOPICS NEED TO BE COVERED? / REFERENCES
maintenance through predictive indications with
immediate positive impacts on plant downtime and
availability. Nevertheless, more powerful on-site and
in-situ characterization tools and techniques as well
as remote automatic failure detection supported
by AI are necessary.
Aerial infrared (IR) thermographic inspections
with drones for PV module quality assessment are
increasingly used at utility-scale level, because of
its high degree of automation, low cost and decent
precision. However, future research should be
focused on its integration not only with existing
SCADA and monitoring systems, but also with
the rest of inspection technologies already being
widely deployed, such as IV-curve tracing, Electroand Photoluminescence imaging. It is desirable,
therefore, that all these streams of data coming
from the field converge into a unified and centralized
digital ecosystem, with AI-analytics at its core,
finally closing the gap between field inspections,
real power losses and maintenance actions.
Finally, future research should not neglect the
important role of field workers and their safety,
allowing them to also fully benefit from IoT and
industry 4.0 technologies. Field technicians equipped
with fully connected smart devices would therefore
be able to generate data streams. This stream is fed
seamlessly into cloud-based analytic engines and
would constantly process and analyze the relevant
field data. Those engines would support the decisionmaking processes necessary to efficiently carry out
the maintenance activities that would secure the
reliable operation of our PV fleet.
11
July 2020
www.etip-pv.eu